A Comparative Study of Na ̈ıve Bayes and k-NN Classifiers on the MNIST Dataset

Overview

This project utilizes Docker to ensure a consistent and reproducible environment for data science work related to water pump analysis. It includes tools like JupyterLab for interactive analysis and visualization.

Prerequisites

• Docker installed on your machine. Get Docker.
• Basic understanding of Docker and command-line interfaces.

Setting Up the Environment

Building the Docker Image

To build the Docker image for the first time, run the build.sh script. This script creates a Docker image named waterpump_project and starts a container named waterpump_container.

./build.sh

Rebuilding the Environment

If you make changes to the environment (such as updating environment.yml) and need to rebuild the Docker image, use the rebuild.sh script. This script will stop and remove the existing container, rebuild the image with the latest changes, and start a new container.

./rebuild.sh

Using the Project

Accessing JupyterLab

After running the build.sh or rebuild.sh script, JupyterLab will be available at http://localhost:8888. The command line will provide a specific token for the session. JupyterLab provides an interactive development environment for working with Jupyter notebooks, code, and data.

Restarting the Same Container

If you just want to restart the container without making any changes to the Docker image (like if you haven't made any changes to your Dockerfile or environment.yml), you can simply start the container again. You can do this with the following Docker command:

docker start -ai nb_knn_container

Project Structure

• Dockerfile: Specifies the Docker image configuration.
• environment.yml: Lists all the Conda environment dependencies.
• build.sh: Script to build and start the Docker container.
• rebuild.sh: Script to rebuild and restart the Docker container with changes.

Adding New Dependencies

To add new dependencies to the project:

1. Update the environment.yml file with the required packages.
2. Run the rebuild.sh script to update the Docker environment.

Contributing

Contributions to the project are welcome. Please ensure that any significant changes are accompanied by corresponding updates in the documentation and Docker environment setup.